Maternal perinatal depression is not independently associated with child body mass index in the Generation R Study: methods and missing data matter Karen A. Ertel a,b, * , Ken Kleinman c , Lenie van Rossem d,e , Sharon Sagiv f , Henning Tiemeier g,h , Albert Hofman h , Vincent W.V. Jaddoe d,h,i , Hein Raat h a Division of Biostatistics & Epidemiology, Department of Public Health, School of Public Health and Health Sciences, University of Massachusetts Amherst, Arnold House, 715 N. Pleasant Street, Amherst, MA 01003, USA b Kellogg Health Scholars Program, Department of Society, Human Development, and Health, Harvard School of Public Health, 677 Huntington Avenue, Boston, MA 02115, USA c Department of Population Medicine, Harvard Medical School and Harvard Pilgrim Health Care Institute, 133 Brookline Avenue, 6th Floor, Boston, MA 02115, USA d The Generation R Study Group, Erasmus Medical Center, Postbox 2040, 3000CA Rotterdam, The Netherlands e Julius Center for Health Sciences and Primary Care, University Medical Center, Utrecht, The Netherlands f Department of Environmental Health, Boston University School of Public Health, 715 Albany Street, Boston, MA 02118, USA g Department of Child and Adolescent Psychiatry, Erasmus Medical Center, Rotterdam, The Netherlands h Department of Epidemiology, Erasmus Medical Center, Rotterdam, The Netherlands i Department of Pediatrics, Erasmus Medical Center, Rotterdam, The Netherlands Accepted 23 May 2012; Published online 10 September 2012 Abstract Objectives: To advance the understanding of the relationship between maternal perinatal depression and child overweight, we used appropriate methodology to account for missing data; incorporated three exposure time points; and included adequate covariate adjustment in a large, sociodemographically diverse sample. Study Design and Setting: We used data from 6,782 motherechild pairs in a prospective population-based study. Maternal depression was assessed with the Brief Symptom Inventory at midpregnancy and 2 and 6 months postpartum. Child height and weight were measured at 36 months of age and converted to body mass index (BMI) z-scores. We compared the complete-case and multiple imputation (MI) analyses. Results: Fully adjusted complete-case models showed a positive association between depression at 2 months postpartum and child BMI z-score (b 5 0.19 [95% confidence interval (CI) 5 0.03, 0.36]; n 5 1,732), and no association between prenatal depression or 6-month post- partum depression and child BMI. Using MI (n 5 6,782), there was no association between perinatal depression and child BMI at any time point. Conclusions: Our study adds evidence that postpartum depression is not associated with child growth across the population in high- income countries. Our results highlight the importance of appropriate handling of missing data, adequate covariate control, and the value of studying the conditions that have produced conflicting evidence regarding perinatal depression and child weight. Ó 2012 Elsevier Inc. All rights reserved. Keywords: Child overweight; Perinatal depression; Postpartum depression; Multiple imputation; Missing data; Developmental origins of health and disease 1. Introduction Depression during pregnancy and in the first 6-months postpartum is common, with period prevalences estimated to be 18% and 14%, respectively [1]. Maternal depression tends to co-occur with other adversities, including poverty, marital dissolution, financial difficulties, and comorbid psychiatric disorders [2]. Maternal depression affects par- enting practices [3e6] as well as cognitive and behavioral outcomes in children [7,8], but it remains unknown if peri- natal depression affects infant growth independent of the social circumstances that tend to cluster with depression. This question is particularly important given the current child obesity epidemic as well as continued concerns with underweight in developing countries. The literature on the relationship between perinatal de- pression and child weight shows conflicting results and is * Corresponding author. Tel.: 413-545-0309; fax: 413-545-1645. E-mail address: kertel@schoolph.umass.edu (K.A. Ertel). 0895-4356/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jclinepi.2012.05.013 Journal of Clinical Epidemiology 65 (2012) 1300e1309